►
From YouTube: 2019-11-12 Crossplane Community Meeting
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
Okay,
the
recording
has
started-
and
this
is
the
November
12th
2019
crossplane
community
meeting.
Let's
go
ahead
and
jump
into
updates
with
the
project
and
milestones
and
releases
and
such
as
the
first
bit
of
news
to
share
since
the
last
community
meeting
is
that
we
were
able
to
release
our
first
ever
patch
release.
A
So
we
have
all
the
release,
engineering
and
automation
in
place
now
that,
with
the
build
automation,
pipelines
and
CI
systems
and
all
that
stuff,
such
that,
when
we
branched
for
these
0.4
release,
we
were
then
able
to
push
more
commits
that
branch
and
test
them
and
when
we
were
ready,
go
ahead
and
roll
out
a
patch
version.
That
is,
you
know,
just
updates
to
that
particular
release,
so
that
enables
master
to
move
on
with
new
features
after
we
branch
for
release.
A
So
when
we're
in
a
code
freeze
or
a
feature
freeze
period
that
are
really
only
be
in
a
released
branch
and
it
will
be
holding
up
master
from
getting
new
features
in.
So
we
are
it's
nice
to
be
able
to
have
that
ability
now
to
release
patches
and
fix
issues
that
our
users
and
community
find
out
in
the
wild
after
we've
already
released
something
so
that
is
good,
and
so
the
next
release
for
work.
Eon
is
0.5
and
that's
actually
your
really
short
release
here.
A
The
intent
is
to
have
a
release
out
right
before
cube
con
san
diego.
We
want
to
be
able
to
put
the
release
out
and
have
some
have
some
noise
and
some
new
things
available
going
into
cube
con,
which
is
less
than
a
week
away.
Q
con
starts
on
Monday
here
in
San
Diego,
so
hopefully
everybody
has
their
travel
plans
and
accommodations
and
such
all
set
up
and
registered
for
the
conference,
because
Monday
morning
everything
gets
going
down
here.
A
So
let's
take
a
look
at
some
of
the
big
things
in
0.5
here,
so
it
is
as
I
mentioned,
it
is
a
short
release.
0.4
was
out
maybe
three
weeks
ago
or
so,
or
two
actually
I
think
so
so
you're,
not
five
is
please
just
under
the
length
of
a
month.
So
there's
not
a
ton
of
new
features
and
new
developments
and
such
that's
going
in
there,
but
there
are
some
interesting
things
that
we
are
focusing
on
and
trying
to
wrap
up
this
week.
A
So
the
first
thing
on
the
list
here
is
that
we
have
been
cleaning
a
integration
of
crossplane
with
get
lab
as
a
get
lab
managed
app
for
for
a
little
bit
now,
and
that
is
all
coming
together
for
get
labs
latest
release
in
12.5,
which
is
nearing
the
end
of
its
lifecycle
now
to
with
feature
freezes
and
try
to
get
everything
immersed
in.
So
basically
when
this
is
complete
here,
the
effort
on
crossplane
side,
which
is
almost
complete,
that
you
will
be
able
to
from
the
get
lab
UI
are
from
the
Atlanta
project.
A
You
have
hosted
there,
you'll
be
able
to
have
cross
playing
provision.
Recent
manage
services
and
resources
for
you
directly
from
your
cat
lab
projects.
Do
you
just
there
Otto,
DevOps
pipelines
and
stuff
to
get
lap
runs.
So
that's
really
excited
to
be
able
to
get
an
opportunity
for
crossplane
to
actually
run
in
a
scenario.
More
realistic
scenario
like
see
ICD
scenario,
where
you're
deploying
your
services
and
your
applications
and
using
cross
plane
to
do
some
of
that
work
for
you.
A
So
super
excited
about
that
and
Steve
is
not
here
today,
actually
because
he's
meeting
with
the
gate
lab
folks
to
try
to
drive
that
to
completion
along
the
same
vein
of
the
CI
CD
stuff
that
we
were
just
talking
about
with
get
lab,
we're
also
focusing
on
integrating
with
other
CI
CD
systems
dan.
Do
you
want
to
give
us
an
update
with
the
guide
and
the
work
that
you've
been
doing
on
integrating
with
at
least
our
ago,
CD
yeah.
B
Absolutely
so
I
think
the
guide
is
pretty
much
ready
to
go
at
this
point
so
for
anyone
who
isn't
familiar,
what
we're
doing
there.
On
the
last
episode
of
tv
s,
we
had
a
demonstration
and
I've
also
added
support
for
AWS.
Now,
essentially,
what
Argo
CD
is
is
basically
just
doing
that
get-ups
provisioning
that
we've
demonstrated
from
you
know
using
jus
control.
It's
just
doing
that
automatically
and
watching
get
repos.
So
when
you
push
changes,
it'll
automatically,
basically
keep
control
apply
any
updates
to
that.
B
But
it's
nice
to
have
that
and
a
user
interface,
and
you
can
set
up
by
targeting
this
cross
CD
repo
that
jared
has
up
here.
You
can
set
up
different
environments
to
already
have
classes
and
networking
in
place
and
that
sort
of
thing
and
then
deploy
applications
into
that
environment.
So
there's
a
PR
open
on
cross
plane
right
now
with
the
Argos
CD
guide,
and
it's
pretty
much
in
a
finish
form.
It
still
needs
a
little
bit
of
review
I
think
before
it
gets
merged,
but
it's
been
tested
and
working
in
that
sort
of
thing.
B
So
that
should
be
a
really
nice
workflow,
especially
for
quickly
testing
things
just
being
able
to
spin
up
environments
really
fast
and
also
getting
sort
of,
like
a
visual
representation
of
that
can
sometimes
illustrate.
What's
going
on
behind
the
scenes,
a
little
bit
more
so
that'll
be
exciting
to
see.
A
Argos
CD
is
pretty
cool
from
what
I've
seen
I
think
it's
a
nice
streamlined
experience
to
be
able
to
you
know,
do
continuous
deployment.
It's
focused
on
you
know:
kubernetes
ecosystem,
but
yeah,
the
visualization
of
all
the
resources
that
it's
deploying
before
you
through
your
pipelines.
That
stuff
is
pretty
cool,
so
whatever
they
like
here
is
that
you
know
0.4.
We
had
a
big
focus
on
being
able
to
enable
these
continuous
deployments,
and
you
know,
get
ops
type
of
scenarios.
A
So
we
laid
the
foundation
and
now
we're
able
to
demonstrate
cross
plane
being
used
in
these
more
realistic
scenarios
at
the
pulling
your
apps
and
your
services
through
continuous
deployment.
So
I'm
super
excited
about
all
that.
We're
also
continuing
our
effort
on
on
maturing
in
progressing
our
various
types
and
controllers
to
the
beta
level
of
stability.
So
we
did
that
for
caches
and
databases
for
GCP
and
AWS
and
0.4,
and
we're
focusing
on
Azure
right
now
as
well
to
get
a
sure
up
to
beta
quality.
Wafak.
A
C
It's
kind
of
like
phew,
if,
like
the
better
quality
high
fidelity
resources
that
requires
real,
like
you
know,
I
have
any
instance
view
at
any
time
you
want,
like
you
know.
If
an
operation
is
ongoing,
you've
gotta
see
the
you
know
in
progress
and
all
that
stuff
Oh
API
is
a
little
bit
different
than
other
resources.
The
database
API,
so
I'm
struggling.
You
know
applying
the
patterns
that
we
know
from
other
providers,
but
it
doesn't
fix
I,
see.
A
A
Markus
is
driving
that
effort
there
and
we
will.
You
know
since
her
top
five
is
a
short
milestone
and
this
effort
is
fairly
significant.
I,
don't
think
that
we're
gonna
have
you
know
we're
not
gonna
have
a
fully
fleshed
out
finalized
security
model
in
0.5,
but
we
are
attempting
to
make
some
strides
in
progress
towards
that
it
within
the
milestone
here.
So
we'll
figure
that
out
before
the
release
and
thank.
A
Luke
ooh,
yeah,
Luke,
Luke
Webber
is
an
amazing
engineer
and
he
writes
great
documents
to
a
lot
a
lot
of
design
coming
out
of
that
boy's
head
right
now,
alright,
so
that
is
kind
of
the
focus
on
0.5
the
high-level
items
there.
The
plan
is
for
0.5
to
be
out
to
be
released
by
the
end
of
this
week.
Cube
con
starts
on
Monday
and
we
would
like
to
release
on
Friday
here
so
that
we
are
going
into
cube
con
with
this
brand-new
release.
A
A
Let's,
we
can
move
on
to
the
community
topics,
section
I
kind
of
talked
about
the
gate,
lab
Auto,
DevOps
integration
that
we've
been
doing
so
that
is
being
wrapped
up
actively
right.
Now
the
the
folks
that
have
been
doing
that
integration
efforts
in
the
crosslink
community
have
have
gotten
the
ins
in
Scenario
working
and
there's
four
different:
merge
requests
that
go
into
different
parts
of
the
get
lab
system
to
get
approved
and
put
into
there.
A
You
know
there's
like
that,
adding
it
as
a
get
lab,
manage
staff
and
then
like
a
database
migration
and
their
Auto
DevOps
home,
chart,
updates
and
documentation.
So
it's
it's
a
bit.
It's
a
bit
of
a
process
to
get
this
new
work
in
to
get
lab
upstream,
but
the
get
lab
team
has
been
really
awesome,
with
getting
feedback
to
us,
providing
suggestions
and
kind
of
shepherding
us
through
the
process.
There
so
definitely
really
really
happy
with
the
effort
from
get
lab
to
enable
crossplane
to
get
integrated
in
for
their
12.5
release.
A
A
So
there
is
this
pull
request
here
that
I
have
linked
in
the
agenda
document
that
Simon
opened
and
we
are
starting
to
take
some
give
us
a
feedback.
Mick
IND
and
I've
already
provided
some
feedback
on
this
pull
request,
but
that's
super
excited
to
see
another
stack
getting
built
by
someone
by
a
you
know,
someone
else
in
the
community
and
ecosystem
to
enable
services
and
they're
on
offering
as
well.
So
it's
not
just
the
big
cloud
providers.
It's
has
a
broader
scope
and
more
reach
than
that
which
is
I'm
pretty
excited
about.
A
B
B
A
So
that's
awesome,
cool,
yes,
and
so
then
just
another
note
that
you
know
coupons
next
week,
and
so
we
I
think
there's
a
fair
amount
of
folks
that
are
going
to
be
in
San
Diego
and
we
are
all
very
friendly
people
and
definitely
interested
in
chatting
with
anybody
about
crossplane.
Or
you
know
your
efforts
there,
your
lessons
in
multi-cloud
and
what
your
pain
points
are
all
that
sort
of
stuff
or
life
in
general.
We're
actually
really
friendly,
so
we'd
love
to
talk
to
people
yeah,
so
hopefully
we'll
get
to
meet
so
meet.
A
Some
new
people
make
some
new
friends
here
in
San
Diego
and
let
me
know
if
anybody
wants
to
go
out
for
a
search,
I
think
there's
gonna
be.
It
looks
like
there's
going
to
be
a
little
small.
Next
week's
time
excited
already
cute
Khan
comes
to
town
and
the
way
has
come
back,
which
actually
there
might
not
be
much
time
for
surfing,
but
we'll
make
it
we'll
make
out
alright.
So
we've
got
a
couple,
pull
requests
that
we
want
to
bring
up
here
and
discuss
in
this
format.
D
Are
yeah
so
this
this
is.
This
is
an
epic
of
an
issue
and
it
all
starts
with
Luke
Weber's
design
changes
around
stack
rolls
and
some
of
the
are
back
concerns
and
the
scoping
of
resources,
and
so
anyway,
his
design
introduced
a
bunch
of
necessary
changes.
I'm
still
I'm,
linking
in
here
some
more
sub
issues
and
creating
some
issues
along
the
way.
D
A
And
then
Marcus
I
took
a
little
pass
through
0.5
milestone
earlier
today
and
I
didn't
like
there's
a
couple
of
perhaps
smaller
scoped
issues
that
might
make
sense
to
put
in
to
address
in
the
remaining
time
in
0.5.
So
take
a
look
at
the
mouse,
the
issues
that
are
in
the
0.5
milestone
now
around
stock
manager
and
see.
If
there's
you
know,
we
could
chat
more
about
like
what
the
priority
should
be,
or
you
know
what
we
can
accomplish
reasonably
before
before
the
end
of
week,
yeah
yeah.
D
A
B
So
I
just
dropped
a
link
in
the
chat
to
the
actual
implementation
of
the
framework,
and
so
essentially,
what
it's
doing
is
allowing
us
to
write
table
different
integration
tests
for
our
controllers
and
CRTs,
and
it's
a
pretty
configurable
system.
It
wraps
the
in
test
package
that
controller
runtime
uses
to
implement
its
test
with
ginko,
and
but
you
know,
we
kind
of
wanted
to
go
with
the
table
driven
test
and
that's
how
we
write
our
unit
tests
and
stuff
and
if
you
actually
go
to
example,
test
two
got
that
go.
B
So
essentially,
what
it
allows
you
to
do
is
reference
a
directory
of
CR
DS
attach
any
controllers
you
want
to
the
manager,
and
you
can
actually
run
this
with
any
cluster
so
into
test
defaults
to
actually
loading
that
CD
binaries
in
the
API
server
binaries,
but
something
I've
been
experiencing
I've
been
experimenting
with,
which
I
think
would
be
really
cool,
is
having
something
like
a
gke
cluster
running.
That's
basically
just
used
for
our
integration
tests
or
even
spending
one
up
using
crossplane
and
and
then
running.
B
These
tests
and
what
you
can
do
is
basically
install
these
CR
DS
go
through
the
steps
of
a
test
and
I've
kind
of
created
these
concepts
of
jobs
and
tests.
So
the
job
would
be
an
environment
that
you'd
want
to
run
tests
in.
So
that
would
be
the
controllers,
the
CR
DS,
any
configuration
of
the
cube
client
there
and
then
run
these
steps
through
it.
B
So,
for
instance,
in
this
example,
you
create
a
namespace,
then
a
GC,
p
provider,
then
a
bucket
class
and
then
another
namespace,
and
it's
just
kind
of
an
example
here,
but
there's
some
concepts
as
far
as
like
cleanup
and
that
sort
of
thing.
So
what
I'm
hoping
to
do
is
get
some
basic
integration
tests
using
this
framework
into
our
stack,
which
should
allow
us
to
move
faster
with
more
confidence
as
we're
implementing
features
more
rapidly
and
I.
Think
it's
still
yet
to
be
determined.
What
it's
going
to
look
like
for?
B
B
The
design
doc
right
now
is
a
little
bit
out
of
date
with
how
the
framework
looks,
but
I've
kind
of
been
going
back
and
forth
with
Nick
reviewing
the
actual
framework
itself
and
then
showing
some
like
examples
of
how
we're
gonna
like
inject
the
config,
the
cue
config
and
that
sort
of
thing,
and
how
we
want
to
do
that.
So,
hopefully,
we
have
some
basic
integration
tests
in
place
by
the
end
of
this
week
that
we
can
run
on
a
regular
basis.
Yeah.
A
I
think
any
effort,
investment
that
we
make
into
integration
tests
to
you,
know
instantly
or
ensure
us
from
regressions
or
from
new
bugs
showing
up.
You
know
that
validation
that
you
get
on
a
continuous
basis
at
a
higher
level
than
just
a
unit
tests
offer
is
always
a
really
good
investment.
I'm
really
happy
to
see
this.
My
biggest
question
is:
how
do
pronounce
this
tin?
It
aphid.
B
B
And
he
kind
of
like
it
was
like
a
playoff
cross
plains.
All
those
things
like
planes
and
stuff
like
that
and
since
we're
wrapping
the
Ames
test,
library
and
affa
did
is
another
word,
for
it
is
a
ramjet
and
it's
basically
a
jet
engine
that
uses
air
propulsion
to
like
propel
itself
forward.
So
it's
kind
of
like
wrapping
around
something
else
and
using
it
to
achieve
even
more
functionality.
So
my.
B
A
A
Yesa
great
thanks
for
all
that
effort
there,
and
so
then
another
poor
quest
I
wanted
to
bring
up
was
the
design
on
parent-child
relationship
labels
or
being
able
to
to
enable
the
scenario
where
given
AC
Rd
like
a
wordpress
instance,
you
could
easily
query
the
entire
kubernetes
the
control
plane
and
find
all
of
the
components
that
are
associated
with
it.
We
we
had
some
feedback.
Some
discussion
on
this
PR,
but
I
think
it's
kind
of
stalled
out
a
little
bit
Steven.
A
Sounds
great
and
yeah
I'll
try
to
dick
Daniels
here,
but
I
would
definitely
once
I
would
like
to
know
what
the
what
you
all
had
run
into
for
not
using
the
common
labels
of
the
upstream
recommended
labels.
So,
let's
we
try
to
get
some
feedback
from
him
on
that
awesome
and
then
one
more
pull
request.
That
is
getting
a
bit
stale
here,
because
the
Daniel
had
started
an
effort
around.
What
is
our
definition
of
done
for
a
given
work
item
or
a
pull
request?
A
What
does
it
mean
to
actually
be
done
with
it
and
begin
a
little
bit
more
formal
about
that
definition
and
that's
kind
of
that
stall
down
a
little
bit
as
well,
but
if
you'd
give
it
got
some
new
feedback
like
four
four
days
ago,
so
hopefully
Daniel
will
be
able
to
update
this
pull
request
with
the
to
incorporate
the
feedback
and
be
able
to
get
get
some
traction
on
this
and
they
get
this
burst
as
well.
Your
sense
you're.
D
A
That's
great
yeah
I'd
love
to
get
that
merged
in
there.
So
we
could
just
point
at
that.
It's
a
smart
alecks
like
Marcus.
We
could
just
point
it.
This
doc
stay
cool,
yes
and
then
Dan
has
dropped
in
the
chat.
That,
after
did,
is
the
short
name
for
arrow,
thorough,
dynamic,
duct,
aha,
very
intuitive,
all
right.
So
that
is
everything
I
had
on
the
agenda
today.
The
floor
is
open
for
anybody
that
wants
to
bring
up
any
community
topics
or
requests
or
anything
that
they
want
to
discuss.
It
was
not
in
the
agenda.
A
A
You,
okay:
if
there
are
not
anymore
agenda
items,
then
we
can
go
ahead
and
adjourn
for
the
week
and
the
next
time
that
we
get
together
we'll
be
down
here
in
sunny
Southern
California
in
San
Diego
at
cute
con,
so
looking
forward
to
seeing
anybody
who's
able
to
make
that
trip
and
be
here
and
yeah.
Let's,
let's
work
on
the
0.5
release
friend
get
that
out
by
the
end
of
week
here
and
keep
this
momentum
going
sweet
thanks.
Everybody.